From Pont d'Avignon to Ponte Vecchio. The Resolution of Constitutional Conflicts Between the European Union and the Member States Through Principles of Public International Law

(1)

Law Department

From Pont d'Avignon to Ponte V ecchio The Resolution of Constitutional Conflicts

Between the European Union and the Member States Through Principles of Public International Law

Christoph U. Sc h m id

LAW No. 98/7

EUI W O R KIN G PAPERS

(2)

European University Institute 3 0001 0032 6581 8 © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(3)

EUROPEAN UNIVERSITY INSTITUTE, FLORENCE DEPARTMENT OF LAW WP F a 9 EUR

&

rs

fc %

Ml

vO

»

EUI Working Paper LAW No. 98/7

From Pont d’Avignon to Ponte Vecchio

The Resolution of Constitutional Conflicts

Between the European Union and the Member States

Through Principles of Public International Law

Ch r i s t o p h U. Sc h m id

BADIA FIESOLANA, SAN DOMENICO (FI)

(4)

No part of this paper may be reproduced in any form without permission of the author.

European University Institute Badia Fiesolana I - 50016 San Domenico (FI)

Italy © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(5)

From Pont d' Avignon to Ponte Vecchio

The Resolution o f Constitutional Conflicts between the European Union and the Member States through Principles o f Public International Law

Dr. Christoph U. Schmid, EUI Florence

"(...) The point of legal institutions is to set up a framework that facilitates and enhances the public use of reason; a structure of discussion which must not be deformed either through external or internal coercion and which allows the continuing of public discourse when shared understandings of lesser generality have broken down."

John Rawls, Political Liberalism, 1993, p. 41.

I. Introduction

In 1974, in the famous "Solange" decision of the Federal Constitutional Court (FCC)', a serious conflict between the EC-Treaty (TEC) and the German Grundgesetz (Basic Law) came to light for the first time. It remains unresolved and to a considerable extent threatens the good relationship between both legal orders and their highest courts. On the German side, the FCC, in its role as the guardian of the Grundgesetz and in line with its jurisprudence on foreign and se curity policy in general,* 2 * claims for itself the exceptional competence to review European law to be applied on German territory in the light of the constitutional "integration clauses" (imposing basic requirements on the participation of Ger many in the EC and the Union). On the contrary, on the European side, the Euro pean Court of Justice (ECJ) has always regarded such a competence as incom patible with its own jurisdiction and, therefore, as a breach of the EC-Treaty. Even though the FCC has raised the threshold conditions for its activation in the Solange II decision,’ it has up to now never ceased to insist on the existence of such a competence.J Now, in the Banana conflict, the competence could actually

* For valuable suggestions I am most grateful to Claus-Dieter Ehlermann, Carol Harlow, Christian Joerges, Karlheinz Ladeur, Agustin Menendez Pelayo, Bruno Simma and Francis Snyder. For the correction of the English version, I wish to thank Richard Burnley.

'BVerfGE 37, 271 - Solange I.

2See H. Schwarz. Die verfassungsgerichtliche Kontrolle der AuBen- und Sicherheitspolitik, 1995. ^BVerfGE 73. 339 (374 et seq.) - Solange II.

“'See the following up decision on the Maastricht treaty in: Entscheidungen des Bundesverfassungsgerichts (BVerfGE) 89, 155 - Maastricht. Among the rich literature on the conflict since the Maastricht judgement see

Bleckmann/Pieper, Recht der In te rn a tio n a l Wirtschaft 1993, 969; Boom, American J. of Comparative Law

(1995), 17; Cremer, Europarecht 1995, 21; Everting, Integration 1994, 165; idem, Gediichtnisschrift fiir Gra- bitz, 1995, 57; idem, CML. Rev. 33 (1996), 401; Fromont, Juristenzeitung 1995, 800; Frowein, Zeitschrift fiir

(6)

be made use of for the first time: The administrative court of Frankfurt has re quested the FCC to review some provisions of the EC banana regime.* 5 which it considers to be at odds with the freedoms of property and profession enshrined in the Grundgesetz.6

In a recent contribution, the relationship between the ECJ and the FCC has been compared to the Cold War logic of "mutual assured destruction".7 Unlike the mere threat to do so, actually setting aside a European act as unconstitutional would be very hazardous since other States could follow this example ("domino effect") relying on some reciprocity rationale, thus putting an end to legal uni formity, which is a basic requirement of the rule of law within the EC. With re gard to the constraints of globalisation,8 this could generate fatal economic and geopolitical consequences even for Germany itself. However, the scenario of the mutual dissuasion should not be viewed in only the negative sense. Thus, the de

auslàndisches òffentliches Recht und Vôlkerrecht 1994, 1; Gersdorf, Deutsches Verwaltungsblatt 1994. 675;

Gòtz, Juristenzeitung 1993, 1081; Grimm, Recht der Arbeit 1996, 66 = Columbia J. of European Law 3 (1996),

229 (English version); Ip sen, Europarecht 1994, 1,9; Heintzen, Archi v fiir Òffentliches Recht 119 (1994), 564;

Herdegen, CML. Rev. 31 (1994). 235; Hirsch, Neue Juristische Wochenschrift 1996. 2457, 2459; Horn, Deut

sches Verwaltungsblatt 1995, 89; Hommelhoff/Kirchhof (ed.), Der Staatenverbund der Europàischen Union, 1994; Kirchhof, Deutsche Richterzeitung, 1995, 253; E. Klein. Gedàchtnisschrift fiir Grabitz, 1995, 271; H. H.

Klein, Maastrichter Vertrag und nationale Verfassungsgerichtsrechtsprechung, 1993; Kokott, Archiv fiir Òf

fentliches Recht 119 (1994) 207; Kònig, Zeitschrift fiir auslàndisches òffentliches Recht und Vôlkerrecht 1994, 17; Koopmans, Nederlands Juristen Blad 1994, 245; Lenz, Neue Juristische Wochenschrift 1993. 3038; Mac

Cormick, Juristenzeitung 1995, 17; Meesen, Neue Juristische Wochenschrift 1994, 549; Oppermann. Deutsches

Verwaltungsblatt 1994, 901; Pfrang, Das Verhàltnis des deutschen Rechts zum Gemcinschaftsrccht nach dem Maastricht-Urteil des Bundesverfassungsgerichts. 1996; Rodriguez Iglesias, Europàische Grundrechte 1996, 125; Schneider, Archiv fiir Òffentliches Recht 119 (1994), 294; Schroder M., Deutsches Verwaltungsblatt 994, 316; Schroder W., Zeitschrift fiir Rechtsvergleichung 1994, 143; Schwarze, Neue Justiz 1994, 1; Sendler, Neue Juristische Wochenschrift 1996, 825; Steinberger, Festschrift fiir Bernhardt, 1995, 1313; Steindorff, Eu- ropàisches Wirtschafts- und Steuerrecht 1993, 341; Streinz, Europaischc Zeitschrift fiir Wirtschaftsrecht 1994. 329; idem, in: Ipsen et al. (ed.), Verfassungsrecht im Wandel, 1995, 663; Tomuschat, Europàische Grundrechte 1993, 489; Voss, Recht der In te rn a tio n a l Wirtschaft 1996, 324; A. Weber, Festschrift fiir Everling, 1995, 1625; Weiler, Jahrbuch fiir Òffentliches Recht 1996, 91 = Festschrift fiir Everling. 1995. voi. 2, 1651 (English version); Winkelmann (ed.). Das Maastricht-Urteil des Bundesverfassungsgerichts, 1994; Wittkowski, Bayeri- sche Verwaltungsblatter 1994, 35 et seq.; Zuck, Neue Juristische Wochenschrift 1994, 978; Zuleeg, Juristenzei tung 1994, 1.- See, for the similar debate in Ireland. Phelan, Revolt or Revolution: the Constitutional bounda ries of the EC. 1997. For an overall view see Fédération International du droit européen (ed.), ‘Le droit consti- tutionel national et l'intégration européenne', 1996.

5Verwaltungsgericht Frankfurt a. M., Europàische Zeitschrift fiir Wirtschaftsrecht 1997, 182 (19 et seq.); for comments see A. Weber. Europàische Zeitschrift fiir Wirtschaftsrecht 1997, 165; P. M. Huber, Europàische Zeitschrift fiir Wirtschaftsrecht 1997, 517; Classen, Juristenzeitung 1997, 454; Vachek, Zeitschrift fiir Rechtsvergleichung 1997, 136.Zuleeg, Neue Juristische Wochenschrift 1997, 1201.

6In this context, the term "constitutional conflict" was first used by Voss, Recht der Internationalen Wirtschaft 1996, 324.- However, the conflict could also be solved on the public international law track, since the WTO- Dispute Settlement Body, in its resolution of 25/9/1997, has found several violations of GATT and GATS. It is to be expected, therefore, that the EC will modify the Banana regime in order to comply with the WTO resolu tion. Furthermore, such resolution should have direct effect in Community law, see Eeckhout, CML.Rev. 34 (1997), 11; C. Schmid, Neue Juristische Wochenschrift 1998, 190.

7Weiler/Haltern, 37 Harvard Int. L. J. (1996), 411,446; Weiler/Haltern/Mayer, W.Eur.Politics, July 1995, 4, 3

et seq.

8See e.g. Scharpf, Problem solving capacities of mulit-level governance, 1997.

(7)

velopment of the ECJ’s human rights jurisprudence would mainly be a response to the challenges by national courts, especially the FCC.9

Even though this scenario of mutual destruction has worked as an expedient up to now, and may even have contributed to the evolution of the human rights juris prudence of the ECJ, 10 this article argues that a Cold War relationship between the highest courts of both sides and the legal uncertainty arising therefrom is to be considered as unacceptable in a highly integrated polity like the EC. Therefore, an attempt will be made to "re-juridify" the conflict in order to avoid negative consequences on integration. First, the positions of both sides will be briefly ex pounded. This will confirm the premise that the conflict has strong structural ele ments, in that it is but a logical consequence of the different views on the rela tionship between both legal orders and the supremacy of EC law. This leads to the insight that a solution is only to be found by means of a theoretical recon struction, which sheds light on the logical well-foundedness of the different premises. This reconstruction reaches the conclusion that the relationship of both legal orders is to be regarded as dualistic. Starting from this basis, the central premises for the resolution of the conflict will be developed: Since there is no hi erarchical sub- or supraordination between both legal orders, both may not try to subject one another, but must respect each other's mutual autonomy. Furthermore, the ECJ must recognise that parts of its premises are inconsistent: Supremacy ends where fundamental constitutional features of the MSS are at stake. In such case, a solution may only be found by cooperation, according to which the two judiciaries have to agree on a constitutional standard (e.g. in human rights protec tion) acceptable for both (here called "concordance solution"). If a concordance solution were to fail, the cooperation between both legal orders must be pursued on a higher, third level, through a conciliation mechanism. Even though the es tablishment of such a procedure is to be left to the European legislator, this article will show that a MS can already claim it de lege lata. This solution will be based on the "concrétisation", by PIL principles, of the EC’s duties of solidarity and of respect for national identities of the MSS enshrined in Arts. 5 TEC and F I TEU.

^However. further ahead, Weiler and Haltern concede the limits of their comparison: There is no "non- proliferation treaty" in EC law, so that several constitutional courts could initiate a Cold War at the same time; moreover, courts are not the only actors - rather, governments might use the threat that their courts could set aside a European act as a tactical weapon in a bargaining process at EC level (37 Harvard Ini. L. J. (1996), 411, 438 et seq.).

/0Generally shared view, see Ipsen. EuR 1994, 1 (9f.): Ossenbiihl. DVB1. 1993, 753 (762). Weiler/Lockhart. 32 Common Market L. Rev. (1993) 579; Bleckniann/Pieper, RTW 1993, 969. 9761.; qualifying this view however

Everting, GS f. Grabitz. 1995, 57 (74). © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(8)

II. The genesis of constitutional conflicts

1. The integration clauses of the Grundgesetz and their control by the FCC

According to the Basic Law, the transfer of state sovereignty to the EC through the German statutes of ratification of the EC treaties is subject to indispensable conditions. In contrast to the new Art. 23 Basic Law, the old Art. 24 Basic Law (which is still relevant for the EC treaties and any secondary law based on them) did not name them explicitly. Rather, they must be defined pursuant to common methodological principles. Thereby, it must first be recognised that these provi sions which give constitutional authorisation to integration may not be regarded as a breach of the constitution, but have to be read in the light of its other provi sions, particularly the so called "eternity clause" of Art. 79 III Basic Law.” This contains a reference to human dignity and the value of human life (Art. 1 Basic Law) as well as the fundamental federal, democratic, social principles on which the Federal Republic of Germany is based (Art. 20 Basic Law) and which may not even be set aside by the constitutional legislator acting by unanimity. In case of conflict, the conflicting principles must be balanced so that each of them re tains a maximum of effectiveness (the so called device of "practical concor dance"12). The new Art. 23 Basic Law, introduced before the ratification of the Maastricht treaty, now incorporates these limits to integration explicitly. There fore, it may be understood as an abstract 'balancing formula' along the lines of the doctrine of practical concordance.

Like any other constitutional provisions, the integration clauses in Art. 23 and former Art. 24 Basic Law have to be monitored by the guardian of the Grundge setz: the FCC. This is a clear obligation of the court from which it cannot dis pense itself. Therefore, in the light of German constitutional law, the FCC did not have any alternative to the exercise of such control over EC law to be applied in Germany. Since direct control is not procedurally provided for, the control can only be exercised in an indirect way, i.e. over the "bridge" of the review of na tional ratification statutes: To the extent that a European act exceeds the limits of the integration clauses, the ratification statute (having the rank of standard law, inferior to the constitution) is void, and as a result, the European act is devoid of legal force on national territory.” As will be shown in more detail, this mecha nism, however, does not in principle affect Germany's obligations under EC law.

” See Streinz, Bundesvcrfassungsrechtlicher Grundrechtsschutz und Europaisches Gemeinschaftsrecht (abridged: Grundrechtsschutz). 1989, 247 et seq.

^2The father of this famous concept is former constitutional court judge Konrad Hesse. See Hesse, Grundziige des Verfassungsrechts der Bundesrepublik Deutschland. 20th ed. 1995, at No. 72, 28.

^ F o r details see Streinz, Grundrechtsschutz, 1989, 247 et seq.

(9)

2. Autonomy and supremacy of EC law

As opposed to Art. 23 Basic Law, the EC treaties did and do not deal with the interface between EC and national law in any way. Rather, all related questions remained to be resolved by jurisprudence. To this end, the ECJ developed over the years the well-known doctrines of autonomy, direct effect, supremacy, pre emption, direct effect of directives, directive-conforming interpretation ("indirect effect") and state liability. They can be reconstructed in terms of a supranational "conflict of laws", relying on the techniques elaborated in private international law.74 As is known, all these doctrines became generally accepted by national courts, albeit after strong resistance by some French and German courts. These doctrines have brought about the "constitutionalisation of the treaties", thereby promoting integration even during years of political stagnation.JS However, two elements of these doctrines turned out to be in potential conflict with national constitutions: the unlimited autonomy and supremacy of EC law.

a) Autonomy

Already in the first years of the EC's existence, the ECJ defined its concept of the autonomy of the EC law:

"The conclusion to be drawn (...) is that the EC constitutes a new legal order of internatio nal law for the benefit of which the states have limited their sovereign rights, albeit within limited fields, and which binds both their nationals and themselves."* 16

In the first place, the autonomy of EC law means that - like any international treaty - the EC treaties are principally not subject to restrictions by the internal law of the contracting parties. This is even true for requirements contained in na tional constitutions. In particular, the sole reference to the need for ratification in Art. 247 TEC can not bring about the reception of such national requirements into EC law. Thus, in the case Internationale Handelsgesellschaft,17 the ECJ stated:

"Recourse to the legal rules or concepts of national law in order to judge the validity of measures adopted by the institutions of the EC would have an adverse effect on the uni formity and efficacy of EC law. In fact, the law stemming from the Treaty, an independent source of law, cannot because of its very nature be overriden by rules of national law, how ever framed, without being deprived of its character as EC law and without the legal basis of the EC itself being called in question. Therefore the validity of a EC measure or its effect within a MS cannot be affected by allegations that it runs counter to either fundamental rights as formulated by constitution of that State or the principles of a national constitu tional structure."

,4See C. Schmid, Vertical and diagonal conflicts in the Europeanisation process, Florence 1997 (forthcoming

as a Commission document).

76Since Les Verts, case 294/82, ECR 1986, 1339, 1365 (confirmed in opinion 1/91, ECR 1-1991, 6084, 6102), the ECJ refers to the treaties as the Community's constitution. As to the concept of constitutionalisation, see

Weiler's classic, The Transformation of Europe, Yale L.J. 100 (1991 ), 2403.

i6ECJ cases 26/62, ECR 1963, 1 - Van Gend & Loos; and 6/64. ECR 1964, 1250 - Costa/Enel. ,7Case 11/70, ECR 1970, 1125. © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(10)

Interestingly, Advocate General Warner advocated the contrary position, the socalled "Hypothekentheorie" (mortgage theory), some years later:

"The court has already said in general terms that is cannot uphold measures incompatible with fundamental rights recognized and protected by the Constitutions of MSS (...) I would be inclined to refine on this and to say that a fundamental right recognized and protected by the Constitution of a MS must be recognized an protected also in EC law. The reason lies in the fact that, as has often been held by the Court (...), EC law owes its very existence to a partial transfer of sovereignty by each of the MSS to the EC. No MS can, in my opinion, be held to have included in that transfer power for the EC to legislate in infringement of rights protected by its own Constitution. To hold otherwise would involve attributing to a MS the capacity, when ratifying the Treaty, to flout its own Constitution, which seems to me impossible (...)"«

However, in the Hauer judgement the Court strongly opposed this view by confirming its former opinion:

"As the Court declared in its judgment Internationale Handelsgesellschaft, the question of a possible infringement of fundamental rights by a measure of the EC institutions can only be judged in the light of EC law itself. The introduction of special criteria for assessment stemming form the legislation or constitutional law of a particular MS would, by damaging the substantive unity and efficacy of EC law, lead inevitably to the destruction of the unity of the Common Market and the jeopardising of the cohesion of the EC."19

This view of the Court has remained unchallenged, and has ever since consti tuted the basis of its jurisprudence.20 * Moreover, the democratic homogenity of the EC and its MSS and particularly the fact that the fundamental principles en shrined in Art. 79 III Basic Law are also acknowledged in the ECJ's jurispru dence do not mean that these principles are relevant to the EC in exactly the shape they have been given by the FCC.2J The fact that the ECJ has often had recourse to national constitutional provisions in order to shape EC law (especially human rights) does not prove the contrary. For, in this sense, national constitu tional law is not a legal source of EC law, but merely a "law-finding source". If EC law depended on the consent of 15 national courts, its uniform and efficient application would finally be impossible.

Beyond that, the ECJ and parts of the literature sustain the proposition that the EC has, either with its genesis or through the process of constitutionalisation of the treaties, emancipated itself from the national ratification statutes, i.e. cut the ties to its basis in public international and national constitutional law.22 Already the statement in Costa/ ENEL in which the ECJ talked for the first time of an

«C ase 7/76, ECR 1976, 1229, 1236 - Irca. «C ase 44/79, ECR 1979, 3727 at 14 - Hauer.

20See recently Case C-473/93, ECR 1996,1-3207, at 37 et seq. - Commission/ Luxemburg.

2JThis view is however held by the German author Eibach, Das Recht der Europaischen Gemeinschaften als Prüfungsgegenstand des Bundesverfassungsgerichts, 1986, 105 et seq.

22See Everling's famous essay; Sind die Mitgliedstaaten der Europaischen Gemeinschaft noch Herren der Ver- trüge?, in: Festschrift fiir Mosler, 1983, 173 et seq. (190); idem, Deutsches Verwaitungsblatt 1993, 936 (942); for a similar view Oppermann, Europarecht, 1991, 196, at No. 525.

(11)

autonomous legal order and not any longer of a special order of international law. could be understood in this sense. In the literature, the "emancipation thesis" was mainly advocated by Ipsen in his famous "Gesamtaktstheorie" (collective act theorie). Thereafter, the genesis of the EC system does not primarily depend on the national ratification statutes as in the case of ordinary PIL treaties. Rather, what is crucial is the German participation in the collective act establishing the EC which is based only on Art. 24 I GG and, for this reason, did not need to comply with the rest of the Grundgesetz, so that any problem of compatibility with it could not emerge at all. In the later jurisprudence, the emanicpation per spective tends to underlie the ECJ's reference to the Treaties as the EC's "constitution", a term traditionally reserved for the "higher law" of a sovereign state. An even stronger indication in this direction can be found in the more recent first EEA-opinion whereafter a certain hard core of the treaties could not even be changed by the European legislator.23 According to this view, the MSS are no longer the uncontested "masters of the treaty"2-', and the EC is becoming a widely autonomous polity, its law being no longer subject to unlimited modification or control by the MSS. As a result, MSS could ensure the EC's respect for their Constitutions' limits to integration only collectively through legislative action, in cluding those Treaty modifications the ECJ would still allow.

As far as the FCC is concerned, it is true that, in its jurisprudence, it has also ascribed to the EC system the quality of an autonomous legal order.25 However, in substance, it denies a complete autonomy of EC law in this sense by maintain ing the competence to control the constitutionality of EC law to be applied inter nally. This would not be possible, if the "bridge" (Kirchhof) of national statutes of ratification over which the judicial control over EC law is exercised by national constitutional courts would be cut. For, then, this bridge would have become a bridge in the Avignon style, on which the States may still wish to dance but which does not lead anywhere any longer.

b) Supremacy

From the autonomous character of EC law, the ECJ has also deduced its su premacy vis a vis national law:

"The integration into the laws of each MS of provisions which derive from the EC and more generally the terms and the spirit of the Treaty, make it impossible for the States, as a corollary, to accord precedence to a unilateral and subsequent measure over a legal system accepted by them on a basis of reciprocity. Such a measure cannot therefore be inconsistent with that legal system. The law stemming from the Treaty, an independent source of law, could not, because of its special and original nature, be overridden by domestic legal pro

23Opinion, 1/91, ECR 1-1991. 6084 at 6102.

24As to the various meanings of this concept see Heintzen. Archiv des Offenllichcn Rechts 119 (1994), 564. 25See BVerfGE 52, 293 (296); 31, 145 (173); 37, 271 (277). © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(12)

visions, however framed, without being deprived of its character as EC law and without the legal basis of the EC itself being called into question („.)"2S

The formula "domestic legal provisions, however framed" clearly shows that the ECJ has in mind a primacy of EC law with respect to any national law, includ ing constitutional law. The ECJ's use of the term 'constitution' seems to confirm this conclusion as well. So, from the ECJ’s perspective, supremacy would encom pass the integration clauses of national constitutions and the limits to integration stated by them. However, if this were entirely correct, the EC could actually dis regard these limits by adopting, under a majority regime, measures contrary to them; once the ECJ would have accepted a measure as compatible with EC law, the MSS would dispose of no other remedy against it. As will be shown in more detail, MSS should however have at their disposal a further device for the judicial protection of their Constitutions. This device will render necessary the limitation, to a certain extent, of primacy of European law over national constitutional limits to integration.

In the Protocol on the Application of the Principles of Subsidiarity and Proportionality an nexed to the Treaty of Amsterdam,26 27 the Council has expressed the view that subsidiarity and proportionality rationales cannot be drawn upon in order to limit the primacy of Euro pean law: "The application of the principles of subsidiarity and proportionality shall respect the general provisions and the objectives of the Treaty, particularly as regards the maintain ing of the acquis communautaire and the institutional balance; it shall not affect the princi ples developed by the Court of Justice regarding the relationship between national and EC law, and it should take into account Article F(4) of the Treaty on European Union, accord ing to which the Union shall provide itself with the means necessary to attain its objectives

and carry through its policies". However, this provision is not meant to limit the compe

tence of control exercised by national constitutional courts by virtue of their constitutions.2S

3. Consequences for the control of EC law by national constitutional courts

The premises of the irrelevance, in EC law, of national limitations to integration and of the unlimited supremacy of EC law are considered to imply that there is no scope for the review of EC law by national courts. Particularly, such a power would seem to be incompatible with the ECJ's position of ultimate umpire as to the interpretation and validity of EC law, as laid down in Art. 164-177 TEC.

This function of the ECJ is indeed indispensable to guarantee uniform application of EC law throughout the EC. Therefore, the reference procedure in Art. 177 TEC is particularly

26Costa/Enel, Case 6/64, ECR 1964, 586.

27See Treaty of Amsterdam, final version, CONF/4007/97, TA/P/d 28, No. 2.

2sFor a different view, see Hasselbacli. Juristenzeitung 1997, 942. However, it should be noted that even an explicit statement by the Community legislator as to an unlimited primacy of European law would not change the constitutional conflict in any way, since its origin in national constitutional law would remain unaffected. Futhermore, German organs would not be allowed to vote for such a provision in the Council, since it would clearly violate Art. 23 Basic Law.

(13)

important, since it ensures that the ECJ is consulted before a potentially divergent interpre tation of EC law by national courts. Within EC law, the ECJ's monopoly of interpretation clearly comprises the review of the Treaty's legislative competence provisions as well.29 From this perspective, if not the review procedure as such, at least the finding of a EC act as unconstitutional by a national court would clearly be in breach of EC law. It could be sanctioned by a treaty violation procedure (Art. 169 TEC) and other devices, including the imposition of a fine (Art. 171 TEC).

III. A theoretical reconstruction of the conflict

The analysis in the last chapter shows that both positions are based on different conceptual premises and reach different results as to the degree of autonomy and supremacy of European law and the control competence of national constitutional courts according to a coherent reasoning. Thus, we seem to face a structural conflict which, by the means of positive law, can only be alleviated, but not re solved in all cases. However, a definitive opinion can only be delivered after a further analysis which transcends the internal perspective of each legal order and allows for an overall assessment in which the premises themselves of the two positions are critically reviewed. Such an analysis must go beyond positive law and resort to a theoretical reconstruction of the conflict. Only such an assessment will finally be capable of discerning the potential of both positions for conflict resolution and of developing, on that basis, a plausible proposal at the level of positive law.

29For a different, but unconvincing view, see Schilling, 37 Harvard Int. L.J. (1996). 389. 406 et seq. and idem, Zeitschrift fur Rechtsvergleichung 1997. 96 et seq. According to Theodor Schilling, it is possible that a court of an international organisation which does not possess legislative Kompetenz-Kompetenz may well be endowed with judicial Kompetenz-Kompetenz. i.e. may well be the ultimate umpire with respect to the interpretation of legislative competence provisions. However, this judicial competence would include the power of deliberately exceeding the legislative competence of the organisation (!). In Community law. however, the existence of such a judicial Kompetenz-Kompetenz of the ECJ would be uncertain: Arts. 173-177 TEC might be relied on in fa vour of it. whereas Arts. 4 and 164 TEC would point against it. Now. this problem could be solved by according the ECJ only a "formal", and not a "substantive" judicial Kompetenz-Kompetenz', as a result, national courts would retain a substantive judicial Kompetenz-Kompetenz, which would ultimately allow them to review the ECJ's decisions on the interpretation of legislative competence provisions. This argumentation seems ill- founded. First, it is true that the Community is an order of limited competences and therefore, does not have legislative Kompetenz-Kompetenz (although one might read another conclusion into the ECJ's first opinion on the European Economic Area. Opinion 1/91. ECR 1-1991, 6079. which would however be ultra vires). Sec ondly, however, it should be clear that the ECJ was given the power, in the interest of legal unity, to decide as the ultimate umpire also on competence issues. This power even includes very wide interpretations of legisla tive competence provisions, which might be considered as ultra vires by some. but. nevertheless, according to normal standards of public international law. are binding unless the transgression is essential and manifest (see, specifically on this problem. Streinz, Grundrechtsschutz, 1989, 324. text at fn. 177). Now. it is wrong to deduce from the competence of (even extensive) interpretation an explicit competence to deliberately exceed the bonds of limited legislative competences. Therefore, it is already not necessary to draw on Arts. 4 and 164 in order to justify a limitation o (such a judicial Kompetenz-Kompetenz. since it simply does not exist. For this very reason, there is no need to differentiate between a formal and a substantive judicial Kompetenz-Kompetenz either. Finally, it seems self-evident that Schilling's view would be prone to destroy legal unity in the Com munity and therefore counteract the intentions of the fathers of the Treaty.- See also the convincing response to Schilling by Weiler/Haltern, 37 Harvard Int. L.J. (1996), 411. 423 et seq. For the origins of the concept of 'Kompetenz-Kompetenz', see Lerche, Festschrift Heymanns Verlag, 1995, 409.

(14)

The point of departure for this reconstruction must be the theoretical alterna tives which exist with respect to the relationship of both legal orders: (1) EC law is only an annex of each national legal order in an overall monistic system; (2) the inverse situation: in a monistic system, EC law is supraordinated over national law, and finally (3) both legal orders stand in a dualistic juxtaposition. Now, for the sake of further analysis, the theoretical devices of a certain school must be chosen. In the present context, the devices of the Pure Theory of Law developed by Hans Kelsen, Adolf Merkel and their successors seems particularly well suited, since it is deliberately confined to the reconceptualisation of empirical phenomena, thus excluding as far as possible normative and especially ideologi cal considerations.30 31 Besides that, the devices of institutional legal positivism, developed by Neil Mac Cormick and Ota Weinberger,37 were used in the litera ture. They will be included in this analysis, too.

1. Important basic concepts of the Pure Theory of Law

The Pure Theory of Law defines as a legal order an order of legal norms which has been recognised as a unity. The unity of this order is due to its deductability to a common reason of existence, a (fictitious) Grundnonn. In this hierarchical structure, it is the highest rule for law-creation, which commands the validity of all subordinated norms. In such a chain of validity, the subordinated norm also commands, besides its other contents, the validity of the norm subordinated to itself. In this conception, the simplest solution for C law would be to derive it from the Grundnonnen of the national constitutions of the founding countries (monism with supraordination of national law). Logically, it is also conceivable that, through a legal "revolution", i.e. the coming into existence of a new Grund

nonn, EC law has emancipated itself from its national law base either right at the

foundation of the EC or later on, and that it has taken its place above national law in a monistic system (monism with supraordination of EC law), or in juxtaposition to the latter (dualism).

However, according to Kelsen, besides the derivability from a common Grund-

norm, a legal system presupposes that the efficiency of its norms is guaranteed

through adequate sanctions. This requirement is based on the fact that a Grund-

norm has a purely heuristic value. Thereafter, the recognition of a Grundnorm is

only possible if a real system actually exists; for law is not created in the moment of its construction by legal science, but is given to it as a product of social and empirical processes.32 * So, what is needed is a constitution which, on the whole, is

30For a Pure Theory of Law reconstruction of supranationality see Grussmann, Grundnorm und Supranation- alitat, Rechtsstrukturelle Sichtweisen der europaischcn Integration, in: v. Danwitz et al. (eds.), Auf dem Wege zu einer europàischen Staatlichkeit, 1993, 47; Richmond, Preventing the Identity Crisis, Law and Philosophie 1997, 377.

31MacCormickJ Weinberger, Grundlagen des Institutionalistischen Rechtspositivismus, 19S5.

32Dreier, Sein und Sollen. Bemerkungen zur Reinen Rechtslehre Kelsens, in: idem, Recht-Moral-Ideologie,

1981,217(222). © The Author(s). European University Institute. version produced by the EUI Library in 2020. Available Open Access on Cadmus, European University Institute Research Repository.

(15)

efficient.-33 Drawing on Merkel, Wolf-Dietrich Grussmann however refutes this reasoning: The efficiency of an order of norms would only be an arbitrary crite rion for the heuristic value of the assumption of its validity.34 Only by foregoing this requirement, a theoretical analysis of old constitutions, with are no longer active at the moment of observation (e.g. Nazi or communist-GDR-law) would be possible.

This view must however be disputed. Law must, at least within the minimal bonds of the effiency criterion, take into account its social backround if it wants to be more than a theoretical castle in the air. In so far, an impure element has to be introduced into the Pure Theory of Law. This does not exclude the recognition of old constitutional orders, which are no longer efficient at present. Rather, their implications and "legacies" on the current legal order may be considered through intertemporal norms of conflict or other devices. The denial of the efficiency at the time of observation does not preclude a former effiency somewhere in the past. Thereafter, in order to recognise EC law as a legal system, its efficiency has to be scrutinised. If it is declined, EC law can only be regarded as a subordinated annex of national law within a monistic overall order. As a consequence thereof, the ECJ would be subordinated to national courts, and their claim of competence to review EC law which is to be applied within a State would be justified.

2. Monism with supraordination of national law?

In two remarkable recent contributions, Marcel Kaufmann35 and Theodor Schilling36 have denied EC law an effiency independent from national law, and therefore validity as an autonomous legal order.37 The supranational legal order would not possess this quality, because it would prove inefficient in hard con flicts. Nearly all constitutional or supreme courts would reserve to themselves a control competence on grounds of constitutionality (resp. other important features of national law). Since the EC does not to a large extent dispose of its own ad ministration and, therefore, has to rely on national administrations sworn to up hold their constitutions, EC law would not be able to impose itself over national law in cases of hard conflict. For this reason, only "nationalised EC law", i.e. EC law compatible with the critera of validity of national law and therefore applied by national administrations, would be efficient. As a consequence thereof, EC law would remain integrated in the chain of validity of the legal systems to which it owes its very existence: the national legal systems. Thus, the relationship of the two legal orders could be qualified as monistic, national law being supraordinated

33Kelsen, Reine Rechtslehre, 2nd. ed. 1960. 196 et seq.

34Grussmann, Grundnorm und Supranationalital. Rechtsslrukturelle Sichtweisen der europaischen Integration,

in: v. Danwitz u.a. (eds.). Aul'dem Wege zu einer europaischen Staallichkeit, 1993, 47 (50) drawing on Merkl, Hans Kelsens System einer reinen Rechtslehre, Archiv des offentlichen Rechts 47 (1921), 171 (179).

35Kaufmann, Der Staat 1997. 521.

36Schilling, Archiv fiir Rechts- und Sozialphtlosophie 1997, 568. 37Kaufmann. Der Staat 1997, 521 (540).

(16)

to EC law. This interpretation would have the further advantage that it allows for the use of the potentials of democratic legitimacy of the nation states for the EC. Beyond that, this view does not overlook the critique that it conceives of EC law as an - not necessarily uniform - appendix of the national legal orders. However, it wants to avoid the threat of splitting by means of the self restraint of national law, ordained in "opening clauses" like Art. 23 GG, which mandate national law to take into account the needs of the EC through a "practical concordance" bal ancing approach. Thus, conflicts of validity between the two orders could be avoided, and conflicts would remain in the sphere of law; the conflict decision would not require a choice between law and non-law. Finally, giving the German constitutional court the responsability for European integration would mean put ting it into good hands.

This view, which regards EC law as "auesseres Staatsrecht" (external state law), does not match the European reality, though. For the strong position of the ECJ as the guardian of Communtiy law, and especially the competences of the EC for sanctions vis-à-vis MSS who disrespect EC law or the fundamental values and principles of the community order (Art. 169-171 TEC and new Art. 7 TEU) guarantee a degree of independent efficiency to the European legal order, which cannot be neglected even in the case of hard conflicts. It is true that it is irrele vant, in this context, that the case of an avoidance of EC law by a national court has never happened so far, and that it is even a duty for national organs under na tional law to avoid it happening (since all national systems provide only for the membership in a stable community in which the rule of law is respected), since these circumstances do not prove that "non-nationalised" community law would be sufficiently efficient. However, it is decisive that, even if this case actually happened, the EC, disposing of the sanctions provided for in Arts. 169 et seq. TEC, would not be without any defense against the disobeyance of its law. Apart from that, EC sanctions against a MS which fundamentally disobeys EC law, and thereby leaves the basis of the EC's self contained regime might, as ultima ratio, even be supplemented by national sanctions of other MSS (who would then act as a sort of trustor for the EC) pursuant to P1L.

Furthermore, a monism with supraordination of national law would presuppose that the EC is in principle vulnerable to any intervention by each MS, without having any defense. If a state decided to give up its constitutional self restraint with respect to the EC, no legal means would exist against this action. Thus, a MS could simply abolish the whole community with respect to itself by repealing the ratification statute; countermeasures according to Arts. 169 et seq. TEC would have to be viewed as "non-law" or at least as per se illegal, since they would not be compatible with the national Gnmdnonn; the EC would only exist by the grace of the MSS, and could not guarantee the essential uniformity of its law. As a result, the MS would have remained completely sovereign - which is a view that, since the last century, is not even held with respect to PIL, where only a monism with supraposition of international law or a dualism is advocated in the

(17)

literature38. This would clearly contradict reality once again. Moreover, by the assumption that the national legal order would be supraordinated to the EC one, the monistic theory blocks the only conflict solution available in practice: an equal cooperation between both orders and their courts. Finally, the argument ac cording to which only a monistic model with supraposition of national law would allow for the use of the democratic legitimacy potentials of the nation states for the EC is misleading in two senses. First, in the normal case of nationalised Community law, these potentials could be made use of according to any model. Then, with respect to the structural pecularities of the EC, the transfer of the le gitimacy of the nation state to the Community level is not even sufficient in order to realise democratic conditions at EC level. Rather, autonomous EC devices must be drawn upon to supplement the nation state based democratic legitimacy of the EC. It would therefore seem overstated making the choice of a monistic model dependent on it.

To sum up, the assumption of a lack of efficiency and monistic subordination of EC law under national law is untenable because of the EC's independent compe tence in imposing sanctions. Having thus recognised the status of EC law as an autonomous legal order with its own Grundnorm, only two alternatives remain: Either EC law is supraordinated to national law at the peak of a monistic struc ture, or the relationship between both legal orders can only be conceived of as dualistic.

3. Monism with supraposition of EC law or dualism?

As stated above, the functional theory adopted by the ECJ postulates the com plete autonomy of EC law from its constitutional and international law base, par ticularly from the national ratification statutes, and an unlimited supremacy of EC law over all national law, which would not be controllable by national courts. By stressing the autonomy of EC law, the Court does however not make a choice between a monistic model with supraordination of EC law or a dualistic model. It must be noted, though, that, logically, the autonomy of EC law in the sense of a juxtaposition of both orders - which would constitute a dualistic model - would not lead to an uncontrollable supremacy of EC law. Rather, such a model would continue to depend on a national norm opening its own legal order to community norms, a norm which could be controlled by constitutional courts like any other national law. Thus, in terms of the German legal order, the continuing validity of the ratification act which, together with Art. 23 GG, provides for its opening to wards EC law, would be required. Supremacy of EC law would only exist pursu ant to this constitutional authorisation.

By contrast, an unlimited and uncontrollable supremacy would logically only be possible within a monistic model, in which national law would be directly subor dinated to EC law. Only then could EC law itself determine its unlimited suprem

38See Verdross/Sinima, Universelles Volkerrecht. 3rd. ed. 1984, 23 et seq.

(18)

acy over national law, and the control of national courts could be excluded. The statements of some German authors seem indeed to go into this direction: Arts. 23 and 24 would open the German legal order for a law from another source which does not depend on the continuing force of the ratification act, since EC law would impose itself pursuant to EC constitutional rules themselves.39 40 Inciden tally, in terms of legal theory, there is no difference between this version of the functional model and a purely federal model, in which state and federal law is also derived from a common Grundnorm.

Facing the choice now between a monistic model with supraordination of EC law and a dualistic model, it must be conceded that the former can hardly be proven in the face of the actual development of the EC and the Union.'0 An emancipation and supraposition at the foundation stage would presuppose that the EC was created by some sort of pouvoir constituant41 as a federal state, or at least as some other form of autonomous polity - what Schilling called the "big bang theory" of EC law.42 * 44 Such a theory is simply not tenable, since the founda tion treaties have been concluded as normal PIL treaties and have been ratified as such by national parliaments. Conversely, with the establishment of a hierarchi cally superior system depriving them of their constitutional identity and sover eignty, the European nation states would have committed a clear and massive violation of their constitutions, which can hardly be implied from their action. Furthermore, as stated, even the ECJ was still explicitly referring to the PIL- character of the treaties in 1962;45 the term constitution was first used in 1986.'-' Thus, Schilling rightly critises the "bing bang theory" as a piece of legal meta physics and an ex-post rationalisation of developments which were neither fore seen nor intended at the foundation of the EC.

So, at best, an evolution towards the supraposition of EC law may have taken place a long time after the conclusion of the treaties through the constitutionalisation process. But this hypothesis must be discarded as well. First, for reasons of legal certainty, one may rightly claim the need for a formal agreement of the MSS for such a far-reaching change of the EC’s status.45 But even if such a formal re quirement were waived, any expression of consent by the MSS would seem to be crucial. There are, however, no such indications. In particular, no emancipation of

" S e e Everling, Feslschrift fur Bernhard, 1995, 1161 (1 174 et seq.), relying also on Ipsen and Frowein.

40Simma, Netherlands Yearbook of International Law 16 (1985). 111 (127); Streinz. Grundrechtsschutz, 1989,

92 et seq.; Toniuschat, Europàische Grundrechte 1993, 489 (495); Blanke, Die dffentliche Verwaltung 1993, 412 (419); Scholz, Neue Verwaltungsrechtszeitschrift 1993, 817 (818); Kenipen, Archiv des Volkerrechts 35 (1997), 275 et seq. The FCC's Maastricht judgement is also based on this view, see BVerfGE 89, 155 at 184, 190, 198 et seq.

41 As to this concept in the Community context see Murswieck, Der Staat 32 (1993), 161.

42Arichiv fur Rechts- und Sozialphilosophie 1997, 570. ■^Case 26/62, ECR 1963, 1 - Van Gend & Loos.

44 U s Verts, Case 294/82, ECR 1986, 1339, 1365; confirmed in opinion 1/91, ECR 1-1991,6084, 6102. 45Tomuschat, Bonner Kommentar zum Grundgesetz, Art. 24 at No. 48.

(19)

the EC has taken place by virtue of customary law since the majority of the con stitutional or highest courts of the MSS do not in principle exclude control over EC law.4® This is not only true for the German constitutional court. In France, the Advocate General has, in the famous Nicolo case before the Conseil d'Etat, even explicitly discarded the Kelsenian model of supraordination of EC law.46 47 The same is also true for most other European constitutional or highest courts.48 Be yond that, the recent evolution from the EC to the Union shows a considerable regression to PIL patterns of intergovernmental cooperation.49 This is particularly true with regard to the structure of the 2nd and 3rd pillar of the Union, which - despite important transfers from the third to the first pillar - has been left unaf fected as such by the Amsterdam Treaty. Finally, also this form of monism has to face the criticism that it renders impossible an equal cooperation between both legal orders and their courts as the only realistic form of conflict resolution.

As to the only logically remaining alternative, that is a dualistic interpretation of the relationship between both legal orders, it has, first of all, the advantage of keeping this way out of the conflict open. More importantly, however, this inter pretation comes closest to the reality of two legal orders, which are visibly sepa rated despite important interconnections and which are protected by two powerful courts, each of them having the last word with respect to all legal issues arising

within each system. Pursuant to the dualistic logic, both legal orders constitute

logically independent systems with two different Grundnonnen. As a conse quence, conflicts of validity between both orders are excluded by definition, whereas "conflicts of obligation" for individuals facing the equally justified claims for validity of norms emanating from the two orders are possible.

The theory of institutional legal positivsm, founded by Neil MacCormick and Ota Wein berger reaches the same conclusion.50 It conceives of law as a system, directed towards a coherent unity of norms, the existence of a legal order being regarded as a sort of "regulatory ideal". Since the actual enforcement of the legal order ultimately depends on legal institutions, founded and directed by norms, it may be qualified as an "institutional normative system". The interface between legal orders is regulated in each system by spe cific criteria of legal validity, as contained in so-called norms of recognition (Hart). An analysis of the interplay of the EC and national legal orders along these lines of institutional positivism first shows that both orders do not constitute one monistic system, in which EC law would be surpaordinated to national law (monism with subordination of EC law is not

46Stremz. Grundrechtsschut/. 1989. I28etseq.

47Rcprinted in EuGRZ 1990. 99.

48As to England see House of Lords in the Factortainc-casc. 3 CMLR (1990). 375 (380); for Spain see Spanish Constilutional Couri. Declaration D 1-7-1992. XXXIII Jurisprudcncia Constitucional 1992, 460 (472). An unlimited supremacy based on a hierarchical supraposilion was only advocated by the Belgian Court de Cassa tion in the case Le Ski (EuGRZ 1975. 308); however, the new Cour d' Arbitrage claims again a limited compe tence of constitutional review, see Brihosia, Applicabilité directe et primauté des traités internationaux et du droit communautaire. Revue Belge de Droit International. 1/1996, 33.

4SStreinz. Europarecht, 3rd ed. 1996. 35 at No. 112.

50MacCormick/ Weinberger, Grundlagen des Institutionalistischen Rcchtspositivismus, 1985.

(20)

even discussed). Mac Cormick states: "[The monistic picture] does not square well with the fact that the effective legislature for the EC is the council of Ministers, whose members are identifiable only by reference to the place they hold according to state-systems of law: so EC powers of legal change and criteria of validity presuppoe the validity of competences conferred by state-systems, but not themselves validated by EC law. More generally, the institutional theory of law insists on a degree of sociological realism, hence is not in the Kelsenian sense a pure theory. From this point of view, it is clear that institutions of state law look to the state legal order for confirmation of their competences, not treating this as contingent upon ulterior validation or legitimation by the EC; while reciprocally EC institu tions look to the foundation treaties as sufficient for their validation. A pluralistic analysis is in this instance, and on these grounds, clearly preferable to a monistic one.”

The dualistic interpretation qualifies the value of the concepts of "(full) auton omy" and "emancipation” of EC law from its base in constitutional and PIL. For, as we know since Triepelsi, the relationship of ordinary PIL, including interna tional treaties, and national law may also be conceived of as dualistic. Thereafter, the juxtaposition of national and international law, the latter possessing its own Grundnorm and thus full autonomy, would be nothing special at all. As a conse quence thereof, the notions of "emancipation" and "full autonomy" are not helpful in legal theory. What matters than is only the différenciation between dualism and monism, and, within the latter, between subordination and supraordination of the legal orders involved.

Now, from the existence of two Grundnormen, some fundamental conse quences concerning the reach of conflict norms ("bridging norms"), like in par ticular the supremacy rule, may be derived. According to the only theoretically possible conception, the primacy norm cannot change the dualistic structure itself - in other words, it cannot bring about a "legal revolution" through the annexion of a legal order under the Grundnorm of another. This means that the supremacy rule can only go as far the national Grundnorm allows it to go; i.e. EC law can only influence the national legal order in so far as national constitutions allow it to do so through their opening clauses. In the case of German law, the admissible degree of influence is, as shown, determined by the interpretation of Arts. 23, 79 III in the sense of a pratical concordance balancing, taking into account the pecularities and needs of the EC. All in all, these provisions render possible a far- reaching, but not unlimited, "intrusion" of EC law into the national legal order. As opposed to a monistic construction with the subordination of EC law, though, the possible influences of national law on EC law are not unlimited either.

The institutional legal positivism advocates a similar restriction of supremacy: "(...) The le gal systems of MSS and their common legal system of EC law are distinct but interacting systems of law, and hierarchical relationships of validity, within the criteria of validity proper to distinct systems, do not add up to any sort of all-purpose superiority of one sys tem over another. It also follows that the interpretative power of the highest decision making authorities of the different systems must be, as to each system, ultimate (...) What

Volkerrecht und Landesrecht, 1899.

(21)

For the concrete possibilités of avoiding and, if necessary, resolving conflicts, important basic implications may be derived from the dualistic model.5-9 First, it entails that there is no hierarchical sub- or supraordination between both legal orders. Thus, both may not try to subject each other, but have to respect their mutual relative autonomy. Furthermore, the ECJ must recognise that parts of its premises are inconsistent: As shown, supremacy ends where "Grundnorm- related" constitutional essentials of the MSS are stake. In such a case, a solution may only be found through cooperation, in which the two judiciaries have to agree on a constitutional standard (e.g. in human rights protection) acceptable for both; this will be called "concordance solution" here. Instead of a blind octroi of supremacy, the colliding principals have to be optimised so that each will retain a maximum of efficiency.* 54 It is the best solution, which can claim a maximum of legitimacy. As will be shown, these principles are not only a command of reason (which could not justify the abandonment of cogent legal premises anyway) or a purely theoretical deduction from the dualistic model. Rather, they can be found in European and national positive law as well. In the following analysis, it will be shown that the FCC and the ECJ have not yet exhausted the existing potential in order to avoid conflict.

If, however, a concordance solution along the lines just desribed were to fail, the legal ressources are, contrary to what is generally assumed, still not ex hausted. For then, if one were not to think that the conflict should be deliberately left open for tactical reasons, one might still try to resort to a PIL approach of conflict resolution, consisting of conciliation, i.e. mediation and arbitration.55 While being primarily legal - and not political - these devices share the specific

feature that their results are only persuasive and not necessarily binding within the involved legal orders. Thus, at the end of the day, a conciliation device might

bring about (only) an intermediary solution between monism and dualism. Moving back now to the level of positive law, the remainder of the article ex plores to which extent these lessons from legal theory can be realised with the help of the existing legal devices in both national and EC law.

IV. Conflict-avoidance capacities of national constitutional law

On the basis of the divergent perspectives of the ECJ and national constitu tional law, the capacities of national law are confined to procedural devices of conflict avoidance. First, a finding of a European act as unconstitutional must be

55For an earlier similar view see Heintzen. AoR 119 (1994), 564 (583ff.). 54See Alexy, Theorie der Grundrechie, 2nd ed„ 78 el seq.; 152.

55 These two notions are often not distinguished in the literature. Here, conciliation is used as a general term encompassing both mediation and arbitration. Whereas mediation, typically meaning a conciliation procedure by a representative of a third state, which may even have its own interest in the case, is more politically ori ented, arbitration consists of the conciliation by a neutral body like a commission, often composed by interna tional lawyers. Thus, the solution to be proposed here is closer to arbitration. However, the general term con ciliation will generally be used here.

(22)

this indicates is that acceptance of a pluralistic conception of legal systems entails acknowl edging that not all legal problems can be solved legally."

4. Implications of the dualist model on a conflict resolution

The theoretical reconstruction permits a critical evaluation of the well- foundedness of the above-mentioned different premises of the ECJ and the Ger man Constitutional Court, which form the basis of the conflict. From this evalua tion, important implications for the resolution of constitutitonal conflicts will be derived, which, in a further step, will be realised with the devices available under positive law as it stands.

First, the reconstruction shows that the ECJ's postulate of an unlimited suprem acy is logically inconsistent from a dualistic or a monistic perspective with supraordination of national law. The only model which would logically allow for such a far-reaching influence on national legal systems, monism with supraordination of EC law, does not correspond to the legal and social reality of the Euro pean order. According to the dualistic model advocated here, the postulate of the FCC of being the ultimate umpire regarding EC law which is to be applied within the national legal order is, in principle, justified. As will be shown, however, this competence in positive law comprises first of all only the competence to examine constitutional incompatibilities, and not to set aside European law without taking any further preliminary steps. Furthermore, under a dualistic model, Mac Cor- mick’s conclusion that not all legal problems can be solved legally is also correct in principle. However, it does not mean that such conflicts can only be solved politically. Rather, a dualistic structure in which "conflicts of obligation" for citi zens emerge on a regular basis would hardly be viable. This is the clear conse quence of Kelsen's famous quotation from the bible, that no one can serve two masters.52

Instead of resorting too early to a political solution, law must first try to find an "intra-legal" solution. Generally, this may be achieved by trying to avoid or, if these efforts fail, to resolve conflicts. In this context, it should be noted that even the resolution of legal conflicts between several legal orders is a genuinely legal task. This should be particularly true for the EC since, in the face of a smaller de gree of political consensus, law in the EC was always more important than poli tics. The "intra-legal" character of the task of conflict resolution is clearly shown, inter alia, by the existence of a whole legal discipline devoted to the resolution of conflicts between different national legal orders: private international law. To sum up, it may be assumed that conflict resolution is a legal task and that the lack of a resolution mechanism brings to light a weakness of the EC system, but does not cast doubt on the "intralegal" qualification of such measures. This means that, in the EC context, all legal devices for conflict resolution have to be exhausted be fore a political solution may be resorted to.

52Matthew VI, 24; see Kelsen. Reine Rechtslehre, 2nd. ed. 1960. 330.